This invention relates to a novel method of generating chlorine or other halogen by electrolysis of an aqueous halide such as hydrochloric acid or alkali metal chloride or other corresponding electrolysable halide.
It has been proposed to conduct such electolysis between an anode and cathode separated by a diaphragm notably an ion exchange membrane wherein the anode, cathode or both are in the form of a thin porous layer of electroconductive material resistant to electrode attack and bonded or other wise incorporated over the surface of the diaphragm.
Similar electrode-membrane assemblies have been proposed for a long time for use in fuel cells. Such cells have been called "solid polymer electrolyte" cells.
Such cells have been used for a long time as gaseous-fuel cells, and only recently have been successfully tested for the electolytic production of chlorine from hydrochloric acid or alkali metal chloride brines.
In a solid polymer electrolyte cell for the production of chlorine, the electrodes usually consist of a thin porous layer of electroconductive catalytic material permanently bonded on the surface of an ion-exchange membrane by means of a binder usually composed of a fluorinated polymer such as, for example, polytetrafluoroethylene (PTFE).
According to one of the preferred procedures of forming the gas permeable electrodes, as described in U.S. Pat. No. 3,297,484 to Niedrach, powder of electroconductive and catalytic material is blended with an aqueous dispersion of PTFE particles obtaining a doughy mixture containing 2 to 20 grams of powder per gram of PTFE.
The mixture, which may be diluted if desired, is spread onto a supporting metal sheet and dried. The powder layer is then covered with aluminium foil and pressed at a temperature sufficient to effect the sintering of the PTFE particles, obtaining a thin coherent film.
After removal of the aluminium foil by caustic leaching, the preformed electrode is applied onto the surface of the membrane and pressed at a temperature sufficient to cause the PTFE matrix to sinter onto the membrane.
After rapid quenching, the supporting metal sheet is removed and the electrode remains bonded on the membrane.
As the electrodes of the cell are intimately bonded on the opposite surfaces of the membrane separating the anode and the cathode chambers, and are not therefore separately supported by metal structures, it has been discovered that the most efficient way to carry and distribute the current to the electrodes consists in resorting to multiple contacts uniformly distributed all over the electrode surface by means of current-carrying structures provided with a series of projections or ribs which, during the assembly of the cell, contact the electrode surface on a multiplicity of evenly distributed points. The membrane, carrying on its opposite surfaces the bonded electrodes, must then be pressed between the two current-carrying structures or collectors, respectively anodic and cathodic.
Contrary to what happens in fuel cells, wherein the reactants are gaseous, the current densities small and wherein pratically no electrodic side-reaction can occur, in the solid electrolyte cells for electrolysis of solutions, as in the particular instance of sodium chloride brines, give rise to problems of difficult resolution.
In a cell for the electrolysis of sodium chloride brine, the following reactions take place at the various part of the cell:
______________________________________ main anodic reaction: 2 Cl.sup.- .fwdarw. Cl.sub.2 + 2e.sup.- transport across 2 Na.sup.+ + H.sub.2 O the membrane: cathode reaction: 2 H.sub.2 O + 2e.sup.- .fwdarw. 2OH.sup.- + H.sub.2 anodic side-reaction: 4 OH.sup.- .fwdarw. O.sub.2 + 2H.sub.2 O + 4e.sup.- main overall reaction: 2 NaCl + 2H.sub.2 O .fwdarw. 2NaOH + Cl.sub.2 ______________________________________ + H.sub.2
Therefore at the anode, besides the desired main reaction of chlorine discharge, a certain water oxidation also occurs, although to an extent held as low as possible, with consequent oxygen evolution. This trend to oxygen evolution is practically enhanced by an alkaline environment at the active sites of the anode, consisting of the catalyst particles contacting the membrane.
In fact, the cation-exchange membranes suitable for the electrolysis of alkali metal halides have a transfer number different from the unit and, in conditions of high alkalinity in the catholyte some of these membranes allow some migration of hydroxyl anions from the catholyte to the anolyte across the membrane.
Moreover, the conditions necessary for an efficient transfer of liquid electrolytes to the active surfaces of the electrodes and for gas evolution thereat require anode and cathode chambers characterized by flow sections for the electrolytes and gases relatively much larger than those adopted in fuel cells.
The electrodes must conversely have a minimum thickness, usually in the range of 40-150 .mu.m, to allow an efficient mass exchange with the bulk of the liquid electrolyte. Because of this requirement, together with the fact that the electrocatalytic and conductive materials constituting the electrodes, particularly the anode, is frequently a mixed oxide comprising a platinum group metal oxide or a pulverulent metal bonded by a binder having low or nil electroconductivity, the electrodes are barely conductive, in the direction of their major dimension.
Therefore a high density of contacts with the collector is required as well as a uniform contact pressure to limit the ohmic drop through the cell and to afford a uniform current density all over the active surface of the cell.
These requirements have been so far extremely hard to fulfill, especially in cells characterized by large surfaces as the ones industrially employed in the plants for the production of chlorine with capacities generally greater than one hundred tons of chlorine a day.
Industrial cells require, for economic reasons, electrodic surfaces in the range of at least 0.5 preferably 1 to 3 square meters or above and are often electrically connected in series to form electrolyzers comprising up to several tens of bipolar cells assembled by means of tie rods or hydraulic or pneumatic jacks in a filter-press-type arrangement.
Cells of this size pose great technological problems as regards producing current carrying structures, that is current collectors, with extremely low tolerances for the planarity of the contacts and such as to provide a uniform contact pressure all over the electrode surface after the assembling of the cell. Moreover, the membrane used in such cells must be very thin to limit the ohmic drop across the solid electrolyte in the cell. The thickness is often lower than 0.2 mm. and rarely more than 2 millimeters and the membrane may be easily ruptured or unduly thinned out in the points whereto an excessive pressure is applied during the closing of the cell.
Therefore, both the anodic and the cathodic collector, besides being almost perfectly planar, must also be almost exactly parallel
In cells of small size, a high degree of planarity and parallelism can be maintained, meanwhile providing a certain flexibility of the collectors to make up for the slight deviations from an exact planarity and parallelism.
In commonly assigned copending United States application Ser. No. 57255 filed on July 12, 1979, now U.S. Pat. No. 4,343,689 there is disclosed a solid electrolyte monopolar-type cell for the electrolysis of sodium chlorine, wherein both the anodic and the cathodic current collector consist of screens or expanded sheets welded onto respective series of vertical metal ribs, which are offset one another, thus permitting a certain bending of the screens during the assembly of the cell in order to exert a more uniform pressure on the membrane surfaces.
In commonly assigned copending U.S. patent application Ser. No. 951,984 filed on Oct. 16, 1978, now abandoned a solid electrolyte bipolar-type cell is described for the electrolysis of sodium chloride wherein the bipolar separators are provided on both sides thereof and in the area corresponding to the electrodes, with a series of ribs or projections.
To make up for slight deviations from planarity and parallelism, the insertion is contemplated of a resilient means consisting of two or more valve metal screens or expanded sheets coated with a non-passivatable material, said resilient means being compressed between the anode-side ribs and the anode bonded to the anodic side of the membrane.
It has been observed however, that both solutions, as proposed in the two cited Patent Applications, entail serious limitations and disadvantages in cells characterized by large electrodic surfaces.
In the first instance, the desired uniformity of contact pressure tends to be lacking, thus giving rise to current concentrations in points of greater contact pressure with consequent polarization phenomena and the related deactivation of the membrane and of the catalytic electrodes; localized ruptures of the membrane and localized mechanical losses of catalytic material often occur during the assembly of the cell.
In the second instance, a very high planarity and parallelism of the bipolar separator surfaces must be provided for; this however requires precise and costly machining of the ribs and the seal surface of the bipolar separator.
Moreover, the high rigidity of the elements entail pressure concentrations which tend to accumulate along the series, thus limiting the number of assemblable elements in a single filter-press arrangement.
As a result of these difficulties a current distributor screen when pressed against the electrode may even leave some electrode areas untouched or contacted so lightly that they are essentially ineffective.
Comparable tests which have been made by pressing distributor screen against pressure sensitive paper capable of showing a visible impression corresponding to the screen have shown that substantial areas ranging above 10 percent to as high as 30 to 40 percent of the screen area produce no marking on the paper. This indicates that such unduly large areas remain untouched.
Applying this observation to the electrodes it appears that substantial electrode surface areas can be inoperative or substantially so.